A Convolutional Neural Network model for stress classification based on electrocardiogram data

Abstract

Several studies have been published in recent years addressing the use of technologies for monitoring and diagnosing mental health. Stress disorders have aroused great interest among researchers and have been widely addressed in several studies due to the negative impacts they can cause to people's quality of life and the possibility of triggering serious illnesses that can seriously compromise physical health. Machine Learning techniques have proven to be a promising alternative to provide technology that can help to deal with challenges in the medical field and Deep Learning models particularly have provided excellent performance when dealing with complex problem domains, such as the classification of types and levels of stress. This study proposes a Convolutional Neural Network (CNN) model for classification of stress through the analysis of electrocardiogram data. The model was trained with a dataset generated from an experiment where individuals were monitored by electrocardiogram while performing stress- inducing tasks. The model achieved an accuracy rate of 97.5% in predictions and presented an adequate balance between performance and computational costs. The results obtained demonstrate the potential of Machine Learning techniques as tools to aid in monitoring and diagnosing mental health. Key Words: machine learning, deep learning, mental stress, electrocardiogram